An Automated Classification of Vehicles and Violation Detection in Special Purpose Lanes
The frequent traffic congestion in major highways, especially through rush hours, led to the strict implementation of special purpose lanes as an efficient way of public transportation. However, due to these special purpose lanes often being uncongested, other vehicles would use this lane to skip tr...
Saved in:
Published in: | 2023 IEEE Region 10 Symposium (TENSYMP) pp. 1 - 6 |
---|---|
Main Authors: | , , , , , |
Format: | Conference Proceeding |
Language: | English |
Published: |
IEEE
06-09-2023
|
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | The frequent traffic congestion in major highways, especially through rush hours, led to the strict implementation of special purpose lanes as an efficient way of public transportation. However, due to these special purpose lanes often being uncongested, other vehicles would use this lane to skip traffic and led to an increase in personnel deployment to monitor the said lane. Current developed automated traffic violation detection systems in the Philippines have included various traffic violation detection but do not include violations for special purpose lanes. Papers that did include special purpose lane violations, however, would use onboard cameras, which might not provide continuous lane monitoring, while others that utilized fixed surveillance cameras had outdated algorithms. Hence, there is a need to develop a system deployed to a fixed surveillance camera setup with updated set of algorithms. Vehicle detection and classification using YOLOv5, and lane detection using a polygonal mask are used to build up the system for violation detection in special purpose lanes. This would be applied to existing stationary traffic surveillance cameras to monitor the roads 24/7. The system developed would aid enforcers in apprehension and would increase the efficiency and objectivity of the apprehension process. |
---|---|
ISSN: | 2642-6102 |
DOI: | 10.1109/TENSYMP55890.2023.10223478 |